Certain embodiments of the present invention relate to a diagnostic ultrasound system which measures and images anatomical structures and their movements. More particularly, certain embodiments relate to methods and apparatus for generating and displaying strain rate signals associated with moving tissue structure by reducing noise due to reverberations and other sources in the strain rate signals.
Within the field of ultrasound imaging, physicians have become interested in using tissue strain and strain rate for clinical measurements. The term xe2x80x9cstrainxe2x80x9d refers to a characteristic of the tissue being examined. For example, the strain associated with muscle tissue corresponds to a ratio of the muscle tissue""s initial length and the change in muscle tissue length during a prescribed time interval. In ultrasound imaging, the rate of change of strain (i.e. strain rate) is typically visually presented to a physician as a colorized 2-dimensional image, where variations in color correspond to different strain rates. It has become apparent that the viability of a segment of the cardiac muscle is related to the amount of muscle strain and temporal behavior of the strain that is performed by, or imposed on the muscle segment. Also, it has been determined that malignant tumors may be detected based on the resistance to compression.
One application of real-time strain rate imaging is in cardiology. The strain rate gives a direct and quantitative measure for the ability of the myocardium to contract and relax. By imaging along the myocardium from an apical view, the local strain rate component along the long axis of the heart may be measured. Measuring the local strain rate component gives information about the local shortening and lengthening of the heart wall. By imaging from the parasternal view, the strain rate component perpendicular to the heart wall gives information about the local thickening of the muscle. Wall thickening measured with M-mode or from the 2D image is a commonly used measure for muscle viability. With strain rate imaging, a direct measure for the thickening is available. The strain rate images may potentially add to the diagnosis of a number of cardiac disorders.
To understand strain rate in more detail, it is assumed that a segment of tissue of initial length Lo may be stretched or compressed or itself lengthens or contracts to a different length L. The one-dimensional strain, defined as                     ϵ        =                              L            -                          L              o                                            L            o                                              (        1        )            
represents a dimensionless description of the change. If the length L is considered to be a function of time, L(t), the temporal derivative of the strain, the strain rate, may be found using the equation                               ϵ          .                =                  δϵ                      δ            ⁢                          xe2x80x83                        ⁢            t                                              (        2        )            
If the velocity, v of every point in the object is known, an equivalent definition of the strain rate is                               ϵ          .                =                              δ            ⁢                          xe2x80x83                        ⁢            v                                δ            ⁢                          xe2x80x83                        ⁢            r                                              (        3        )            
The equations also provide a useful description of the deformation of the tissue segment. The strain rate measures the rate of the deformation of the segment. If the strain rate is zero, the shape of the segment is not changing. If the strain rate is positive, the length of the segment is increasing, and if the strain rate is negative, the length of the segment is decreasing.
Strain rates that occur above a given level are assumed to be non-physiological and, therefore, are artifacts caused by reverberations or other noise sources during imaging. Reverberations are caused by multiple reflections within the tissue. The reverberations and noise may bias the velocity gradient estimated within the tissue due to correlation with a false or corrupted echo. Falsely increased, decreased or even reversed strain rate estimates may result. There is a certain range of physiological strain rates in both normal and diseased tissue. In the normal and diseased human cardiac muscle, for instance, the peak positive and peak negative longitudinal strain rates have been reported as +3.14 sxe2x88x921 and xe2x88x921.78 sxe2x88x921 respectively. The values include artificially increased contraction by stress echocardiography.
One possible explanation for the increased strain rate caused by stationary reverberations is described. Assuming a constant spatial velocity gradient, the velocity samples along a beam line are increasing along the beam. The strain rate may be estimated as the difference between pairs of velocity samples divided by the distance between them, and, in this case, yields a spatially constant strain rate.
Furthermore, assuming a region influenced by a reverberation, there may be a bias brev in the velocity estimate. The effect on the strain rate is a reverberation bias, brev/(2ds), above and below the reverberation region where ds is the distance between two points of the tissue segment. Depending on the amount of the velocity bias due to the reverberation, the estimated strain rate may achieve values outside the normal range.
Previous efforts to solve the problem of noise in the strain rate estimation have used a clutter filter on the tissue velocity data prior to calculation of the strain rate. There are various problems with the method. First, the clutter filter may introduce its own velocity estimation bias, which again is reflected as a strain rate estimation bias. Second, when the velocity is near zero (as in the diastolic diastasis period of the cardiac cycle), the clutter filter tends to increase the variance of the velocity estimates, and thus also the variance of the strain rate estimates.
U.S. Pat. No. 6,099,471 to Torp et al. is directed to a method and apparatus for real-time calculation and display of strain in ultrasound imaging. Ser. No. 09/432,061 (now issued U.S. Pat. No. 6,352,507 B1) to Torp et al. is directed to a method and apparatus for providing real-time calculation and display of tissue deformation in ultrasound imaging.
A need exists for an approach to filtering out non-physiological high strain rates in strain rate imaging due to reverberation and other sources of noise.
An embodiment of the present invention provides an ultrasound system for generating and displaying filtered strain rate signals corresponding to structure within a subject in response to complex Doppler signals. Various combinations of several processing techniques are employed including filtering out high strain rate signals due to reverberation and other sources of noise, complex autocorrelation, velocity signal estimation, real strain rate signal estimation, complex strain correlation signal estimation, complex signal averaging, and real signal averaging.
Apparatus is provided for generating, filtering, and displaying strain rate signals from signals sampled by the ultrasound system. xe2x80x9cFilteringxe2x80x9d, as used throughout, means extracting or modifying those signals that are corrupted due to reverberation or noise. The apparatus includes a Doppler processing module and a strain rate. processing module for performing various combinations of several functions that include complex autocorrelation, velocity signal estimation, real strain rate signal estimation, complex strain correlation signal estimation, complex signal averaging, and real signal averaging.
A method is also provided for generating, filtering, and displaying strain rate signals from signals sampled by the ultrasound system. The method includes filtering out high strain rate signals due to reverberation and other sources of noise, performing various combinations of several functions that include complex autocorrelation, velocity signal estimation, real strain rate signal estimation, complex strain correlation signal estimation, complex signal averaging, and real signal averaging.
Certain embodiments of the present invention afford an approach to generating and displaying filtered, color strain rate images having reduced noise and improved image quality.